Semiconductor Wafer Defect Analysis

Background

Silicon wafers go through multiple fabrication processes such as photolithography, etching, and deposition. At the end of these processes, wafers are inspected for defects such as scratches, particles, and pattern misalignment.

To ensure the manufacturing process remains stable, engineers monitor the number of defective wafers in each production lot. Statistical process control tools such as the np control chart help detect unusual variation in the defect rate. ## np Chart Control Limits

For an np chart:

  • \(n\) = sample size
  • \(\bar{p}\) = average fraction defective

The center line is:

\[ CL = n\bar{p} \]

The upper control limit is:

\[ UCL = n\bar{p} + 3\sqrt{n\bar{p}(1-\bar{p})} \]

The lower control limit is:

\[ LCL = n\bar{p} - 3\sqrt{n\bar{p}(1-\bar{p})} \] ## R Analysis

library(qcc)
## Package 'qcc' version 2.7
## Type 'citation("qcc")' for citing this R package in publications.
defects <- c(8,7,11,3,6,11,4,4,5,8,8,6,7,6,10,7,9,5,6,8,
             8,10,4,11,4,11,6,6,9,5,4,7,5,7,8,5,10,9,7,9,
             4,9,4,7,11,8,10,8,7,5)

n <- 200

qcc(defects, type="np", sizes=n)

## List of 11
##  $ call      : language qcc(data = defects, type = "np", sizes = n)
##  $ type      : chr "np"
##  $ data.name : chr "defects"
##  $ data      : num [1:50, 1] 8 7 11 3 6 11 4 4 5 8 ...
##   ..- attr(*, "dimnames")=List of 2
##  $ statistics: Named num [1:50] 8 7 11 3 6 11 4 4 5 8 ...
##   ..- attr(*, "names")= chr [1:50] "1" "2" "3" "4" ...
##  $ sizes     : num [1:50] 200 200 200 200 200 200 200 200 200 200 ...
##  $ center    : num 7.14
##  $ std.dev   : num 2.62
##  $ nsigmas   : num 3
##  $ limits    : num [1, 1:2] 0 15
##   ..- attr(*, "dimnames")=List of 2
##  $ violations:List of 2
##  - attr(*, "class")= chr "qcc"

Interpretation

The np control chart displays the number of defective wafers in each lot of 200 wafers inspected. The center line represents the expected number of defective wafers based on the average defect rate across the 50 lots.

All observations fall within the control limits and there are no clear patterns indicating special cause variation. This suggests that the photolithography process was operating in a stable and statistically controlled state during the observed period.

These control limits can now be used to monitor future production lots and detect unusual increases in defect rates. ## Complete R Code

library(qcc)

defects <- c(8,7,11,3,6,11,4,4,5,8,8,6,7,6,10,7,9,5,6,8,
             8,10,4,11,4,11,6,6,9,5,4,7,5,7,8,5,10,9,7,9,
             4,9,4,7,11,8,10,8,7,5)

n <- 200

qcc(defects, type="np", sizes=n)

This is an R Markdown document. Markdown is a simple formatting syntax for authoring HTML, PDF, and MS Word documents. For more details on using R Markdown see http://rmarkdown.rstudio.com.

When you click the Knit button a document will be generated that includes both content as well as the output of any embedded R code chunks within the document. You can embed an R code chunk like this:

summary(cars)
##      speed           dist       
##  Min.   : 4.0   Min.   :  2.00  
##  1st Qu.:12.0   1st Qu.: 26.00  
##  Median :15.0   Median : 36.00  
##  Mean   :15.4   Mean   : 42.98  
##  3rd Qu.:19.0   3rd Qu.: 56.00  
##  Max.   :25.0   Max.   :120.00

Including Plots

You can also embed plots, for example:

Note that the echo = FALSE parameter was added to the code chunk to prevent printing of the R code that generated the plot.